Forecasting pedestrian evacuation times by using swarm intelligence

Many models have been developed to provide designers with methods for forecasting the time required for evacuation from various places under a variety of conditions. Particularly for high traffic buildings or buildings of cultural, governmental, or industrial importance, it is of paramount importance to properly evaluate and plan for the necessary evacuation time. To address this need, a number of models for pedestrian simulation, either considering the system as a whole or studying the behavior and decisions of individual pedestrians and their interactions with other pedestrians, have been developed over the years. In this work, a model for evacuation simulation and for estimating evacuation times is proposed. It is inspired by the so-called Particle Swarm Optimization (PSO). The multi-agent-based simulation characteristics of PSO and the way this technique combines individual and collective intelligence make it suitable for this problem. The PSO-based model presented here allows for assessment of the behavioral patterns followed by individuals during a rapid evacuation event. Evaluation of these behaviors can address a variety of public safety concerns, such as architectural design, evacuation protocol definition, and regulation of public space.